Minimally invasive coronary artery bypass grafting

Transcription

1 Catheter-Assisted Totally Thoracoscopic Coronary Artery Bypass Grafting: A Feasibility Study Gilles Soulez, MD, Michel Gagner, MD, Eric Therasse, MD, Fadi Basile, MD, Ignacio Prieto, MD, Philippe Pibarot, DVM, PhD, Claude Laflamme, MD, Louis Lamarre, MD, and Hani Shennib, MD Departments of Radiology, Surgery, Experimental Surgery, Anesthesiology, and Pathology, Hotel-Dieu de Montreal Hospital, and Department of Surgery, McGill University, Montreal, Quebec, Canada Background. The purpose of this study is to examine the feasibility of performing totally thoracoscopic internal mammary-to-coronary artery bypass grafting with the assistance of radiologically guided catheter intervention. Methods. Fourteen dogs were subjected to mobilization of the internal mammary artery and anastomosis of it to the left anterior descending coronary artery over an angiographic catheter inserted into the internal mammary artery under fluoroscopy. The anastomosis was completed over the catheter using sutures and the application of fibrin glue. Eight animals underwent the anastomosis after their sacrifice. The other 6 animals were put on closed chest cardiopulmonary bypass and had their anastomosis done after intraaortic balloon occlusion and cardioplegic arrest of the heart. All animals had an angiographic and pathologic examination at the completion of the anastomosis. Results. Anastomosis was completed in all dogs. Three anastomoses leaked and two were noted to be stenosed at completion of the anastomosis. One leak was sealed by application of fibrin glue. Both stenotic anastomoses were caused by suturing of the back wall when a short angiographic catheter could not be positioned across the anastomosis. Conclusions. Minimally invasive totally thoracoscopic mammary-to-coronary artery bypass grafting with catheter assistance is feasible. Technical improvement and appropriate instrumentation are required to minimize anastomotic failure. (Ann Thorac Surg 1997;64: ) 1997 by The Society of Thoracic Surgeons Accepted for publication April 4, Address reprint requests to Dr Shennib, The Montreal General Hospital, 1560 Cedar Ave, Suite L9-120, Montreal, Que, H3G 1A4, Canada. Minimally invasive coronary artery bypass grafting (MICABG) has emerged as a promising technique for the surgical revascularization of patients with selected coronary artery disease. Currently the most widely performed MICABG procedure is done through a left anterior minithoracotomy, anastomosing the internal mammary artery (IMA) to the left anterior descending artery (LAD) [1 3]. The superiority of the long-term patency of the left IMA (LIMA) and its association with patient longevity is a sound reason to pursue MICABG based on LIMA-to-LAD anastomosis [4 7]. Meanwhile, revascularization of coronary artery disease by catheter interventions such as angioplasty and stenting continues to provide efficient and least invasive methods for revascularization of coronary artery disease. The anatomic location of the coronary artery lesions, their complexity, recurrence of stenosis, and patient choice may dictate the need to resort to surgical revascularization as a primary or secondary technique of choice [4]. The wide acceptance of MICABG would depend on its simplicity, safety, and long-term effectiveness. Because it is designed to be patient friendly, smaller and less painful incisions and avoidance of cardiopulmonary bypass are desired. Although it is currently performed through a variety of incisions such as the anterior thoracotomy, parasternal, or partial sternal incisions, unquestionably it would be desired to perform it through even smaller incisions such as port access using endoscopic instruments. The objective of this study is to explore whether MICABG can be performed totally endoscopically with the assistance of fluoroscopy and catheter intervention. This study reports our preliminary observations on this approach. Material and Methods Animal care and experimentation were conducted in accordance with the guidelines of the Canadian Council on Animal Care, and the protocol was approved by the institutional animal care committee. The study was designed as two parts. The objective of the first part was to examine the feasibility of totally endoscopic dissection of the IMA, percutaneous catheterization of the IMA under fluoroscopic guidance, exposure of the targeted area of the LAD, and performance of the IMA-to-LAD anastomosis over a catheter passed into the LAD from the IMA. This part of the study addressed technical issues relating to the facility of catheter intervention and an approach combining it with totally endoscopic surgical IMA-to-LAD anastomosis on 8 dogs. The second part was performed on 6 living dogs that under by The Society of Thoracic Surgeons /97/$17.00 Published by Elsevier Science Inc PII S (97)

2 Ann Thorac Surg SOULEZ ET AL 1997;64: TOTALLY THORACOSCOPIC CABG 1037 went cardiopulmonary bypass (CPB) and cardioplegic arrest during the course of catheter-assisted endoscopic IMA-to-LAD anastomosis. Part 1 Eight mongrel dogs, 20 to 25 kg each, were anesthetized with midazolam and fentanyl, followed by endotracheal intubation. The right IMA (RIMA) was accessed under fluoroscopic guidance through a right axillary approach in the first 2 animals. Subsequently, in the next 6 animals, right femoral artery access was used because of its simplicity. A 7F guiding catheter (Nycomed, Paris, France) was inserted by the Seldinger technique. A 4F H1 catheter (Cordis, Miami, FL) was advanced coaxially into the RIMA (Fig 1). In dogs, the superior mediastinum is joined bilaterally, so both LIMA and RIMA were accessible from the left side of the chest. The RIMA was chosen for logistical reasons related to limited operating room space and crowding of radiology equipment, radiologist, and surgeon. The animals were placed in the dorsal decubitus position and three 12-mm trocars (Endo-pass; Ethicon, Cincinnati, OH) were inserted in the left fourth and sixth intercostal spaces and the subxiphoid supradiaphragmatic position. The RIMA was dissected under total thoracoscopic vision with a 10-mm, 30-degree angled thoracoscope. The RIMA was mobilized totally from the xiphoid to just below the first rib using cautery dissection. Larger branches were clipped using medium-sized titanium clips (Ethicon). The distal IMA was clipped and transected after heparinization (50 units/kg intravenously). The distal IMA was withdrawn from the thorax through the thoracoscopic trocar. A partial vascular clamp was applied to the proximal artery and the distal end of the artery was trimmed with microscopic scissors and cut at a 45-degree angle under direct vision in preparation for anastomosis. After adequate preparation, the distal IMA was reintroduced into the thorax. The pericardium was incised anteriorly using laparoscopic scissors (Ethicon). Thereafter, the dogs were sacrificed with a bolus of 100 mg of pentobarbital. The mid-lad was exposed and 0 silk sutures were used to snare the LAD proximal and distal to the targeted LAD arteriotomy site. The proximal LAD was clipped with a hemostatic clamp. An incision, 8 mm in length, was done using a size 11 laparoscopic scalpel and microscissors (Ethicon). The distal IMA and LAD were approximated and 6-0 Prolene suture (Ethicon, Somerville, NJ) was used to sew the posterior side of the IMA to the LAD in a running fashion (Fig 2A). A inch guidewire (Terumo, Tokyo, Japan) was then introduced from the IMA to the LAD and the 4F coaxial catheter was advanced over the wire into the LAD across the anastomosis. The toe, anterior wall, and heel of the anastomosis were then completed over the catheter, assuring patency of the anastomosis and the distal run off (Fig 2B). At completion of the anastomosis, fibrin glue (GRF biological glue; FII, Saint-Just, Malmont, France) was applied over the circumference of the anastomosis (Fig 3). Fig 1. (Top) Access to left femoral artery for cardiopulmonary bypass, right carotid artery for aortic clamping and cardioplegia infusion, and right femoral artery for internal mammary artery (IMA) and left anterior descending artery catheterization. (Bottom) Right and left venous and left femoral arterial access for cardiopulmonary bypass. (CCA common carotid artery; EJV external jugular vein; IVC inferior vena cava; SVC superior vena cava.) At completion of the procedure, the catheter was retracted into the IMA and three injections of 5 ml of iothalamate (Conray 60; Mallinckrodt, Pointe-Claire,

3 1038 SOULEZ ET AL Ann Thorac Surg TOTALLY THORACOSCOPIC CABG 1997;64: Fig 2. (A) Right side of anastomosis suture. (B) Left side of anastomosis suture over the angiographic catheter. (IMA internal mammary artery.) Quebec, Canada) were given in the anteroposterior, 45-degree right anterior projections. Fluoroscopic control and intraoperative angiograms were taken with a mobile fluoroscopy unit (BV22; Philips, Eindhoven, the Netherlands). A stenosis was defined as a more than 30% Fig 3. Endoscopic view of an anastomosis completed with the angiographic catheter still inside the internal mammary artery and the left anterior descending artery. Fibrin glue was applied to seal the anastomosis. reduction of lumen diameter. Autopsies were subsequently performed for direct and histologic assessment of the anastomosis and myocardium. Part 2 The 6 animals in the second part of the study underwent IMA-to-LAD anastomosis with the same surgical protocol (part 1); however, this was accomplished on live animals with CPB and cardioplegic arrest. In brief, two 20F venous cannulas (USCI, Billerica, MA) were introduced into the vena cava through two external jugular vein cut-downs (see Fig 1). Arterial access was achieved using a 16F arterial cannula (USCI) inserted into the left femoral artery (see Fig 1). Aortic occlusion and cardioplegia were achieved by the insertion of a 40-mm occlusive latex balloon on an 8F catheter (Meditech; Boston Scientific, Watertown, MA), which was inserted through the right carotid artery into the aortic root (see Fig 1). Inflation of the balloon and occlusion of the ascending aorta was ascertained under fluoroscopic guidance. Proximal injection of iothalamate (Conray 60) assured aortic occlusion and the appropriate positioning of the balloon above the coronary ostia. Subsequently, standard cardioplegic solution was infused through the tip of the catheter. Adequate CPB flow (mean L/min) and cardioplegia was achieved in all animals; however, ventricular distention during aortic clamping was noted in 2 animals. Results In part 1, completion of the anastomosis was achieved in all animals. The anastomoses were tight and patent in 5

4 Ann Thorac Surg SOULEZ ET AL 1997;64: TOTALLY THORACOSCOPIC CABG 1039 between 95 and 180 minutes (median, 150 minutes). Episodes of ventricular fibrillation requiring a second administration of cardioplegic solution occurred in 3 dogs. Fig 4. Angiographic control of a tight and patent anastomosis. dogs (Fig 4). Angiographic assessment revealed two immediate leaks and one stenotic anastomosis. The dog with a stenotic anastomosis had a calibration of the anastomosis only on a inch wire because the angiographic catheter was too short to be positioned across the anastomosis. On pathologic examination, this stenosis was related to an incidental sewing of the posterior wall of the LAD with one stitch. Through the femoral approach, cannulation of the IMA was easy as well as the maneuvering of the distal tip of the catheter across the IMA and to the LAD. In the second part of the study, all animals survived to completion of the anastomosis, permitting its inspection grossly and angiographically. One minute anastomotic leak was observed, which was sealed promptly by the application of fibrin glue and gelfoam. Another anastomosis was noted to be stenotic angiographically and again in this dog the calibration of the anastomosis was undertaken only over a inch wire due to the shortness of the angiographic catheter. On pathologic examination, the stenosis also was related to overbridging of the anterior and posterior wall with a stitch. The time for catheter and percutaneous CBP installation ranged from 45 to 120 minutes (median, 55 minutes). Anastomosis time ranged from 46 to 120 minutes (median, 55 minutes). It was clear that performing the anastomosis over the catheter was simple and permitted completion of the anastomosis with little concern for the inability to visualize the posterior wall and toe of the anastomosis. Endoscopic suturing instruments currently available in this experiment were cumbersome, resulting in breaking of the sutures in 3 of 6 anastomoses. In this experiment, aortic cross-clamp time was long, ranging between 56 and 120 minutes (median, 76 minutes). Cardiopulmonary bypass time was also prolonged, ranging Comment Currently there are two schools of thought on the performance of MICABG. The first and more popular one advocates doing the surgical revascularization on the beating heart with segmental stabilization of the target coronary artery. The most common target conduit is the LIMA-to-LAD anastomosis through an anterior minithoracotomy [1, 2]. Access to more proximal segments of the LIMA for the purpose of total mobilization requires video assistance or increasing the retraction and tension on the chest wall, resulting in more trauma and postoperative pain. Stabilization of the LAD could be achieved mechanically [2] or with a suction device [8]. Although these techniques appear to be safe, there is a question as to how much less invasive they are when compared with the conventional technique through a sternotomy, and no data from a properly randomized study to compare the two approaches are available. Ultimately, it would be theoretically superior if both IMA take-down and its anastomosis to the LAD could be performed totally endoscopically. The second school of thought advocates performing MICABG through small incisions and using CPB through peripheral cannulation (port access). Recently, preliminary reports of clinical application of port-access coronary artery bypass grafting have shown technical difficulties to complete a true port-access procedure. Prolonged operative time led to a modification adopting a minithoracotomy through the bed of the fourth costal cartilage [9]. This approach was based on the results of an experimental model of port-access LIMA-to-LAD anastomosis on percutaneous CPB and with cardioplegic arrest [10, 11]. In this model, Stevens and associates [10] reported a graft failure due to back wall suturing. It is noteworthy that in our study both cases of anastomotic stenosis involved animals that had no calibration of the anastomosis over the angiographic catheter. Calibration devices to anastomose the mammary coronary artery anastomosis have already been employed to avoid incorporation of the IMA back wall during suturing [12]. Use of catheter-assisted totally endoscopic coronary artery bypass technique, as advocated in this study, has the following advantages: (1) We believe that insertion of a catheter into the IMA, particularly if in the future it can be illuminated, may facilitate the take-down of the IMA. (2) Passing the firm catheter across the anastomosis into the distal LAD, if maintained in a good position, may facilitate the performance of the anastomosis with little concern for inducing stenosis or occlusion. (3) At completion of the anastomosis, it is possible to retract the catheter and obtain immediate completion angiography, assuring the adequacy of the anastomosis before closure. There are several disadvantages of this technique: operating rooms need to be updated to permit the

5 1040 SOULEZ ET AL Ann Thorac Surg TOTALLY THORACOSCOPIC CABG 1997;64: collaboration of an interventional radiologist, cardiologist, and cardiac surgeon. The instruments and machines necessary for catheter assistance will need to be available. Cannulation of the IMA may theoretically have the disadvantage of intimal injury rendering it of lesser quality than if not manipulated from within. As we had no midterm evaluation of the IMA after the procedure we cannot presume whether intimal hyperplasia of the IMA will be a significant problem. Further studies with midterm angiographic and histologic evaluation may be required to answer this question. Performance of the anastomosis itself endoscopically is tedious and unreliable, and our time of anastomosis is comparable with that reported in Stevens and associates [10] experimental model (60 26 minutes (mean standard deviation). We can expect that improvement in instrumentation and surgical abilities to manipulate endoscopic sutures will considerably reduce anastomosis time. But despite the presence of a catheter across the anastomosis, the technique of suturing currently used by the totally endoscopic approach is time consuming. It will also probably be necessary to investigate other ways of performing endoscopic anastomosis. The use of a lowpowered CO 2 laser to achieve welding anastomosis could be an alternative. Basu and associates [13] anastomosed the IMA and coronary artery with a CO 2 laser in a canine model. This approach significantly reduced anastomosis time when compared with conventional sutures and could easily be used in our endoscopic model. The development of an automatic stapler and gluing material would be another alternative. These approaches may be appropriate if we use a catheter as described in our model to approximate and calibrate the anastomosis. Our preliminary study indicates that catheter-assisted totally thoracoscopic coronary artery bypass is feasible, if appropriate instruments are available. The evolution of MICABG will undoubtedly continue with emphasis on rendering the procedure less invasive. Furthermore, we believe that it is likely that MICABG will advance from performing single-vessel to multivessel coronary artery bypass and hybrid revascularization by collaboration between cardiac surgeons and interventional cardiologists or radiologists. The setting of specialized angiography-operating room facilities may render a future combined approach more efficient by permitting the simultaneous work of both teams. This study was supported by a grant from the Sterling-Winthrop Institute (No 3031). We thank Suzanne Carioto for invaluable assistance and animal care management. References 1. Benetti FJ, Ballester C. Use of thoracoscopy and a minimal thoracotomy, in mammary coronary bypass to left anterior descending artery, without extracorporeal circulation. Experience in two cases. J Cardiovasc Surg 1995;36: Shennib H, Lee A, Akin J. Safe and effective method of stabilization for coronary artery bypass grafting on the beating heart. Ann Thorac Surg 1997;63: Calafiore AM, Di Giammarco G, Teodori G, et al. Left anterior descending coronary artery grafting via left anterior small thoracotomy without cardiopulmonary bypass. Ann Thorac Surg 1996;61: Kirklin JW, Akins CW, Blackstone EH, et al. Guidelines and indications for coronary artery bypass graft surgery. A report of the American College of Cardiology/American Heart Association Task Force on Assessment of Diagnostic and Therapeutic Cardiovascular Procedures (Subcommittee on Coronary Artery Bypass Graft Surgery). J Am Coll Cardiol 1991;17: Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal-mammary-artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314: Kouchoukos NT, Wareing TH, Murphy SF, Pelate C, Marshall WG. Risks of bilateral internal mammary artery bypass grafting. Ann Thorac Surg 1990;49: Fiore AC, Naunheim KS, Dean P, et al. Results of internal thoracic artery grafting over 15 years: single versus double grafts. Ann Thorac Surg 1990;49: Borst C, Jansen EWL, Tulleken CAF, et al. Coronary artery bypass grafting without cardiopulmonary bypass and without interruption of native coronary flow using a novel anastomosis site restraining device ( Octopus ). J Am Coll Cardiol 1996;27: Reitz BA, Stevens JH, Burdon TA, St. Goar FG, Siegel LC, Pompili MF. Port-access coronary artery bypass grafting: lessons learned in a phase I clinical trial [Abstract]. Circulation 1996;94(Suppl 1):I Stevens JH, Burdon TA, Peters WS, et al. Port-access coronary artery bypass grafting: a proposed surgical method. J Thorac Cardiovasc Surg 1996;111: Schwartz DS, Ribakove GH, Grossi EA, et al. Minimally invasive cardiopulmonary bypass with cardioplegic arrest: a closed chest technique with equivalent myocardial protection. J Thorac Cardiovasc Surg 1996;111: Hickey TJ, Cooper M, Rose EA. Facilitated internal mammary anastomoses. J Thorac Cardiovasc Surg 1992;104: Basu S, Corrado PM, Matthews SC, et al. Internal mammary coronary artery anastomosis by CO 2 laser: an acute experimental study. J Cardiovasc Surg 1991;6: